Decided to stay now that Lynx has sorted some law and order about the place. My thanks to Lynx. First test with the new kit finds it working 100%, current is down to 200ma (not 20ma like i said in the vid lol)
Diode is working, power adjustment working and good voltage on the cell but with no gas yet (I now know why)

https://www.youtube.com/watch?v=RTbp0D_5Khc

Matt, it took me a while to suss it but I know how to bias the modulations at around 300vdc per 3 second burst and it's looking good.

https://www.youtube.com/watch?v=vv013MLJhlg

I can change that parameter to a much quicker bias just by increasing the frequency and the bias on the modulation. I feel real close to getting there at the moment guys.

I can't tell you how good it feels to see you sharing again Nav :-)
Looking forward to see some havoc being wreaked on Big Oil, this small planet needs it.
Humanity needs it.
760 V on the cell with no bubbles, that's just astounding.
Looking forward to see some proper DC bias ringing those bubbles crazy the Pucharich way.
Absolutely love it.
Thanks for sharing mate :thumbsup:

Quote from Lynx on July 27th, 2018, 02:31 PM

I can't tell you how good it feels to see you sharing again Nav :-)
Looking forward to see some havoc being wreaked on Big Oil, this small planet needs it.
Humanity needs it.
760 V on the cell with no bubbles, that's just astounding.
Looking forward to see some proper DC bias ringing those bubbles crazy the Pucharich way.
Absolutely love it.
Thanks for sharing mate :thumbsup:

The reason there is no gas is because no voltage is left on the cell, I could double that voltage to 1500v quite easily but it wouldn't help. It is what you leave on the cell after each set of resonant peaks that matters. It is down to the input signal, each batch of pulses peaks at 768v then resonates down to zero which is no good. What you need to do is send 768v into the cell then only take back 350v at resonance into the choke. So you set the resonance to be biased on both sides of the dc offset with an AM signal so that each 180 degree cycle cannot finish at the dc offset but half way there are 350v, the diode rectifies the lower 180 degrees and it latches onto the top 180 degrees at 350v starting point.
This means that on the signal I showed the dc bias will increase by 350v every 3 seconds which is 7000v per minute. The above signal is a test signal so imagine it at resonance! Once this thing is humming the voltage will increase exponentially at the resonant frequency.

Quote from nav on July 27th, 2018, 02:53 PM

The reason there is no gas is because no voltage is left on the cell, I could double that voltage to 1500v quite easily but it wouldn't help. It is what you leave on the cell after each set of resonant peaks that matters. It is down to the input signal, each batch of pulses peaks at 768v then resonates down to zero which is no good. What you need to do is send 768v into the cell then only take back 350v at resonance into the choke. So you set the resonance to be biased on both sides of the dc offset with an AM signal so that each 180 degree cycle cannot finish at the dc offset but half way there are 350v, the diode rectifies the lower 180 degrees and it latches onto the top 180 degrees at 350v starting point.
This means that every 3 seconds on the signal I showed the dc bias will increase by 350v every 3 seconds which is 7000v per minute. The above signal is a test signal so imagine it at resonance! Once this thing is humming the voltage will increase exponentially at the resonant frequency.

Hubba hubba.
Love it.
Keep it up mate :thumbsup:

You got the idea Nav.  The gotcha I'm running into is the stinking transformer ALWAYS level sets the bias.  Why?  Because induction is the dv/dt of the signal, which means the DC bias gets stripped off of the derivative.  Pucharich found a novel way to get around this and you are darn close Nav to making it work.

Right.

If we jump back in time and think about a water capacitor, it's possible to get one plate to stick at a particular charge level, for at least a small moment.  So what we need is a ping-pong effect between the plates.  Charge one, let it stick, charge the other and before the first one loses its charge, we recharge it again, but this time at a different potential.  Modulation of the wave is clearly the way to do this.  Personally, I don't think one plate always has to be positive and the other plate negative.  We're not interested in separating the Hydrogen from the Oxygen.  So the plates can toggle their relative potential back-n-forth, just not quickly as in a regular A/C wave.  That's Tau at work holding the potentials apart long enough for gas to form.

Man, am I ever looking forward to seeing how you get this going.  It's right there.  You're so close.

Here we see a modulated square wave Stan Meyer style with its trailing edge well above the DC offset. Imagine this signal rectified through a diode, the leading edge of the rectified signal will start well above the DC offset and so the exponential voltage rise is born.

https://www.youtube.com/watch?v=bJh1ZE06fKs&feature=youtu.be

Bad news, I was sweeping frequencies this morning and I head a crack and the system went dead. Pulled the chokes out and the negative coil has blasted away inside and gone open circuit on me as well as frying the diode. You get near to where you wanna be and it all goes pair shaped again. Those cokes were resonant to design specs too which wasn't easy to do. LOL

Which means you could have hit some sweet spot resonating the circuit to bits then.
Not having built something like this myself, would it be a good idea to insert safeguards of sort to the individual building blocks?
For instance fast acting fuses here and there springs to mind.

What you did Nav is something Ronnie warned me about.  He absolutely insisted you need to tune at very low voltage.

The resonant-rise potential with the number of turns on these coils can fry them with ease.  Once you have the wire insulation pitted with tiny holes, it's game over.  Russ did the same thing with his monster Newman coil.

OK I fixed the coil, stripped it down and it had blown where the 31 gauge comes out of the spool at the side and is soldered to fatter gauge wire. That must be its weak point where its not protected by the magnetic field as much, it blew a tiny hole in the hot glue covering it. Matt before things went wrong I noticed something odd. On a test signal similar to the one on the last vid, the DC voltage completely disappeared from my circuit but the AC remained. Switch the cell off with the circuit running and it comes back at 350v. The cell is eating the bias or causing the circuit to eat the bias.

There's no capacitance.

My study so far comes up with conclusions that I found out in other set ups too. The AC bypasses the diode and i've had 4 different diodes now and this latest high speed diode is rated at 1200v breakdown voltage. I think you need a full bridge rectifier if you want to develop a DC bias and keep the AC out. The reason I say this is because the AC circuit defuses the DC circuit and you can't stop it with a single diode. So....an idea would be to treat the cell and the choke output as if you were regulating a power supply, using the cell as a smoothing cap to absorb AC ripple effect.

For Lynx.

Thanks for sharing, most interesting :thumbsup:
That's an easy enough circuit to put together, all it takes is a little willpower, the rest should be easy enough to find in any fuelcell tinkerer's workshop.
If at resonance the AC hits the cell while being biased through the core then the core itself should hold the answer to this phenonema then as the diode should do it's job in rectifying the (amplified) AC voltage.
Do you have that famous gap in the core?
I can recall having read somewhere that the gap should either be directly between the chokes or smack on the opposite side halfway between the core ends.
What if the step up transformer needs to be on the same core as the chokes?
Your schematic is really down to earth elegant in the basic most simple way, dealing only with the very heart of the Pucharich/Meyer way of feeding a water fuel cell using coils around a, I'm assuming, ferrite core then.
This should be used as a motivational piece for experimenting with when ot comes to making some HHO the unconventional way, it's easy enough to put together.

Quote from nav on July 28th, 2018, 10:32 AM

The AC bypasses the diode and i've had 4 different diodes now and this latest high speed diode is rated at 1200v breakdown voltage.

Diodes are funny things.  Without current flow, they do not act like a valve.  I would bet money at the point when the diode "should" switch-off and block, there is no current there, so the diode just stays switched on.

I learned this when building my plasma spark generator.  Take the current away and the diode is useless, but create a current loop for the diode to work against and things start happening.

Quote from Matt Watts on July 28th, 2018, 02:30 PM

Diodes are funny things.  Without current flow, they do not act like a valve.  I would bet money at the point when the diode "should" switch-off and block, there is no current there, so the diode just stays switched on.

I learned this when building my plasma spark generator.  Take the current away and the diode is useless, but create a current loop for the diode to work against and things start happening.

Interesting :thumbsup:

So in theory then a high Ohmic resistor, say 100 Mega Ohm or the likes, would do just fine then connected between the diode's cathode and the choke wiper arm in Nav's schematic.

Quote from Matt Watts on July 28th, 2018, 02:30 PM

Diodes are funny things.  Without current flow, they do not act like a valve.  I would bet money at the point when the diode "should" switch-off and block, there is no current there, so the diode just stays switched on.

I learned this when building my plasma spark generator.  Take the current away and the diode is useless, but create a current loop for the diode to work against and things start happening.

You could be correct there. I think its the way my chokes are configured, they block current at 1990hz and only allow voltage to pass but how can you have LC resonance without normal current? In a tank circuit the voltage of the cap builds up as the rate of current in the inductor increases until the voltage reaches that of the supply (415v in my case). When that happens the voltage in the choke decreases and so does the rate of current. If it were just a tank circuit with a diode then the voltage on the cap would rise to twice the supply voltage which is what mine does, it reaches 768v and stays there which is exactly what the math says it will do minus system losses.The diode stops the choke current from being negative, when zero current is reached across the choke then the system stops charging and stays that way and the cap will be stuck at twice the input voltage for ever.

Quote from Lynx on July 28th, 2018, 04:41 PM

Interesting :thumbsup:

So in theory then a high Ohmic resistor, say 100 Mega Ohm or the likes, would do just fine then connected between the diode's cathode and the choke wiper arm in Nav's schematic.

You were quick to spot that but that's already been tried with several different value resisters. It changes the tone of the ringing in the choke so effectively it slightly changes the resonant frequency of the choke but not to a massive degree, only by a few hz. The problem lies with what Matt is saying I think, the nature of the choke is to stop current flow and i'm stopping current flow at the chokes self resonant frequency of 1990hz, the LC resonance between the positive choke and the cap is hindered because current cannot escape the choke and hit the diode. During testing, the choke will not produce any voltage at the cap either side of resonance. At 4000hz the voltage is zero and at 1000hz the voltage is zero, voltage starts to come in at 1800hz and dies after 2100hz when the ringing of the choke stops. When the choke rings at peak voltage of 768v the current at the diode is zero so there is no continued LC resonance. This means that my chokes are resonant at the wrong frequency, they are resonant at the carrier frequency of the AM signal (1990hz) but will not allow harmonics and modulations to pass.
There may be another way to do it with these particular chokes. If I make the modulations 1990hz then the choke will be resonant at the modulation frequency rather than the carrier frequency and then switch the carrier to 3980hz but keep the TAU as specified as a third part of the signal which I can do. The idea is to choke current but not choke the current needed for LC resonance.

Quote from nav on July 29th, 2018, 03:43 AM

You were quick to spot that but that's already been tried with several different value resisters. It changes the tone of the ringing in the choke so effectively it slightly changes the resonant frequency of the choke but not to a massive degree, only by a few hz. The problem lies with what Matt is saying I think, the nature of the choke is to stop current flow and i'm stopping current flow at the chokes self resonant frequency of 1990hz, the LC resonance between the positive choke and the cap is hindered because current cannot escape the choke and hit the diode. During testing, the choke will not produce any voltage at the cap either side of resonance. At 4000hz the voltage is zero and at 1000hz the voltage is zero, voltage starts to come in at 1800hz and dies after 2100hz when the ringing of the choke stops. When the choke rings at peak voltage of 768v the current at the diode is zero so there is no continued LC resonance. This means that my chokes are resonant at the wrong frequency, they are resonant at the carrier frequency of the AM signal (1990hz) but will not allow harmonics and modulations to pass.
There may be another way to do it with these particular chokes. If I make the modulations 1990hz then the choke will be resonant at the modulation frequency rather than the carrier frequency and then switch the carrier to 3980hz but keep the TAU as specified as a third part of the signal which I can do. The idea is to choke current but not choke the current needed for LC resonance.

Would a full wave rectifier bridge do the job or would that leave unwanted "extra" DC pulses through the chokes?

Quote from Lynx on July 29th, 2018, 03:55 AM

Would a full wave rectifier bridge do the job or would that leave unwanted "extra" DC pulses through the chokes?

Its hard to find bridge rectifiers that run at higher frequency so you'd have to make your own and then you are into big bucks because 4 diodes at the rating as the one I fried would set you back about 100 bucks.
I've just replaced the one in my circuit so i'm into 50 bucks already in diodes.

OK, we have a little gas on tube set C which is the only set switched on, not a lot but gas produced with capacitance not current. I can prove its capacitance because on the lower video I show the lack of gas on my flat plate capacitor in the same experiment, after about 15 minutes there wasn't one bubble of gas.
Now we have to ask ourselves why the gas is not coming out of the gap between the tubes and why only at 27hz?
I think the answer lies in my tube design. Right at the bottom of the tubes there is a dielectric spacer an inch long that sits between the outer and inner tube and I think somewhere around that region the reactances are equal between the positive choke and the tube set C. In other words, the capacitance of the water itself as a dielectric layer was too open circuit and if its too open circuit then the reactance will be wrong.

https://www.youtube.com/watch?v=uAjRTY6fdlI&feature=youtu.be

Flat plate

https://www.youtube.com/watch?v=JOX3sVcrfxc&feature=youtu.be

It is interesting to note that gas does not appear on tube set C until about 10 minutes of it being switched on as if it's conditioning the water in some way.

You'll need to go to youtube and switch to full screen to see the tiny gas bubbles.

Beautiful :thumbsup2:

As for diodes, google showed me DSEI30-12A, 40nS recovery time, 1200V, 26A, TO247, if it's anything near what you're looking for.
https://www.tme.eu/gb/details/dsei30-12a/tht-universal-diodes/ixys/

The gas is actually coming from the negative plate at the bottom but nothing is coming from the positive which could indicate low current electrolysis of the worst kind.

I think my problem lies in the amp. It doesn't produce enough current and although 415v is present at the secondary going into the chokes, there isn't enough current behind it to operate the chokes and diode properly and without current in the choke then the voltage to the cell is limited to low choke current. Time for a bigger amp.